Visual Query Language

Abstract ：
Many applications archive a large number of variables over time, but much of the data in the resulting large databases is never searched, because of the limitations and difficulty of doing so using traditional database query languages.VQL provides a graphical user interface to time-series data, freeing the user who wants to search for matching patterns from the burden of programming in a database query language, or using other labor-intensive techniques. This makes the searches accessible to more people, and saves them time and money in the process. VQL presents the data in graphical form, and the user
simply selects the pattern of interest by dragging the cursor - VQL then searches for matches according to the closeness of match desired. VQL is particularly effective when searching very large databases or real-time streams containing large amounts of noisy data.

Advantages: 
Location of time-series patterns in large amounts of historical or real-time data A visually-oriented approach to analysis of data sequences. The desired time-shape is selected interactively by the user or defined as a template The search engine uses various proven algorithms to find similar shapes and returns a list of matches Search parameters control the flexibility of the search -easy control of false positives The final list of matches is sorted by a fitness function The technology has been proven in practical applications over several years

Applications:
VQL technology is useful in a wide range of applications, such as research, medical (e.g., patient monitoring), building controls, industrial automation,condition monitoring, network traffic analysis, and financial modeling. It can be used in two ways: as a stand-alone application, or within the context of another data-analyzing software program. VQL has been used successfully in several domains, notably Space Station data, energy consumption data, and aircraft engine performance. VQL can be used to search for matches of process data relating to a disturbance or undesirable event in the process. This makes post-mortem analysis much easier, because conditions leading to similar events can then be compared to see what the common precursors are. These precursors can then be eliminated, or used to set up condition monitoring so that early warning can be provided in future. Similar approaches can be used in network traffic analysis, patient monitoring, energy usage in buildings, financial trading, etc.